Better Representation of the Pedestrian Environment in Travel Demand Models

Kelly Clifton, Portland State University


  • Robert Schneider, University of California, Berkeley
  • Jennifer Toole, Toole Design Group


The tools available to transportation planners to forecast non-motorized travel behavior are woefully inadequate for that purpose. Despite recent attention paid to the importance of physical activity for health purposes and transportation policy initiatives that seek to reduce auto use and encourage walking, cycling and transit, extant modeling tools suffer from a lack of spatial acuity and behavioral sensitivity to the preferences of non-motorized travelers. Accurate prediction of the likely responses of travelers to land use changes, parking management, pricing and other policies that would encourage non-motorized travel and thereby reduce emissions also requires a nuanced representation of the pedestrian travel environment. There is a need for analytical modeling tools that can predict likely traveler responses at a more micro level of detail, including travel that has been up to now obscured by the TAZ systems used in travel demand modeling systems. This would seem to be critically important for assessing the impacts a given land use that might be considered as an attractor of pedestrian travel, whether it be a mixed-used development proposal or a new transit station. There is also a need for a tool that could predict the catchment area of such facilities. Perhaps more fundamentally, there would seem to be no analytical model of pedestrian behaviors that can gauge traveler preferences and the tradeoffs they are willing to make between distance and other attributes that might describe the quality of the walking environment. This project aims to fill this gap by developing more robust pedestrian planning tools for use in regional travel demand models. This applied research will improve the mode choice capabilities of Metro\'s existing trip-based model with respect to pedestrian trips. The research design will make use of existing data resources from recent regional household travel surveys, pedestrian count data and built environment attributes to develop better measures of the pedestrian environment that will result in better model performance. This project is a partnership between the Oregon Modeling Collaborative, Metro and Toole Design Group. The project has value in its direct application to Metro\'s upcoming planning efforts as well as the possible integration into other trip-based demand models in other urban areas in the country. It builds on the principle investigators\' previous and current work in non- motorized model development.

Project Details

Project Type:
Project Status:
End Date:
August 31,2013
UTC Grant Cycle:
OTREC 2012
UTC Funding: